Graduate Theses, Dissertations, and Problem Reports

2007

Finite element analysis of slope stability

Pinar Wanstreet
West Virginia University

Follow this and additional works at: https://researchrepository.wvu.edu/etd

Recommended Citation
Wanstreet, Pinar, "Finite element analysis of slope stability" (2007). Graduate Theses, Dissertations, and
Problem Reports. 4347.
https://researchrepository.wvu.edu/etd/4347

This Thesis is protected by copyright and/or related rights. It has been brought to you by the The Research
Repository @ WVU with permission from the rights-holder(s). You are free to use this Thesis in any way that is
permitted by the copyright and related rights legislation that applies to your use. For other uses you must obtain
permission from the rights-holder(s) directly, unless additional rights are indicated by a Creative Commons license
in the record and/ or on the work itself. This Thesis has been accepted for inclusion in WVU Graduate Theses,
Dissertations, and Problem Reports collection by an authorized administrator of The Research Repository @ WVU.
For more information, please contact researchrepository@mail.wvu.edu.
 Finite Element Analysis of Slope Stability

By
Pinar Wanstreet

Thesis Submitted to the

College of Engineering and Mineral Resources
at West Virginia University
in partial fulfillment of the requirements
for the degree of

Master of Science
in
Civil Engineering

Hema J. Siriwardane, Ph.D., P.E., Chair

Udaya B. Halabe, Ph.D., P.E.
Roger C. Viadero, Ph.D.

Department of Civil and Environmental Engineering

Morgantown, West Virginia

2007

Keywords: Slope stability, soil nailing, factor of safety, finite element analysis
 ABSTRACT

Finite Element Analysis of Slope Stability

Pinar Wanstreet

Slope stability is an important aspect of geotechnical engineering. The use of

finite element analysis of slope stability has gained popularity in recent years due to its
capability to handle complex problems. The primary focus of this research was to study
the influence of soil nailing on the factor of safety of stability of slopes by using finite
element analysis, and to investigate failure mechanism. In this paper, stability of various
soil slopes was analyzed using the Shear Strength Reduction (SSR) technique. Finite
element analysis was performed on both reinforced and unreinforced soil slopes and the
results were compared with results from various traditional methods. Finite element
results show that analysis of stability of slopes using the SSR technique is a useful
alternative compared to traditional methods, especially when geometry is complex.
 ACKNOWLEDGEMENT

I would like to express my sincere appreciation to Dr. Hema Siriwardane for

giving me the opportunity to pursue this degree under his guidance and support. His
timely appreciation of my work was a great source of encouragement. His expertise in
geotechnical engineering and finite element modeling improved my research skills and
prepared me for future challenges. I thank Dr. Udaya Halabe and Dr. Roger Viadero for
accepting to serve in my committee and for their participation.

I would like to thank my fellow graduate student Raj Gondle for his help and
sharing his valuable knowledge for my research study.

I would also thank to my husband Josh for his continuous support, care and
patience during my Masters study.

Finally, thanks to my family for their endless love, support and care. Their
encouragement is much appreciated.

iii
 TABLE OF CONTENTS

ABSTRACT ……………………………………………………………………………. ii
ACKNOWLEDGEMENT ……………………………………………………………. iii
TABLE OF CONTENTS …………………………………………………………….. iv
LIST OF TABLES …………………………………………………………………… vii
LIST OF FIGURES ………………………………………………………………….. viii
CHAPTER 1: INTRODUCTION ……………………………………………………. 1
1.1 Introduction …………………………………………………………………. 1
1.2 Types of Slope Failures and Instability Mechanism ………………………... 2
1.3 Conventional Techniques to Improve the Stability of Slopes ………………. 4
1.4 Factor of Safety ………………………………………………………………8
1.5 Problem Statement …………………………………………………………... 8
1.6 Scope of Work ………………………………………………………………10
1.7 Research Objectives …………………………………………………………10
1.8 Previous Studies ……………………………………………………………..11
CHAPTER 2: LITERATURE REVIEW……………………………………………. 12
2.1 Introduction …………………………………………………………………12
2.2 Shear Failure ………………………………………………………………...12
2.3 Mohr–Coulomb Model Description ………………………………………...13
2.4 The Shear Strength Reduction (SSR) Technique ……………………………13
2.5 Determination of Factor of Safety (FOS) …………………………………...15
2.6 Slope Stability with Soil Nails ………………………………………………17
2.6.1 Historical Background of Soil Nailing ……………………………18
2.6.2 Reinforcing Mechanism in Soil Nailing …………………………..18
2.6.3 Soil Nail Installation ………………………………………………20
2.6.4 Nail Heads …………………………………………………………22
CHAPTER 3: NUMERICAL METHODOLOGY………………………………… 26
3.1 Introduction ……………………………………………………………….. 26

iv
 3.2 Finite Element Method (FEM) ……………………………………………. 26
3.3 Geometry ………………………………………………………………….. 27
3.4 Soil Model …………………………………………………………………. 28
3.5 Nail and Nail Head Model ………………………………………………….30
3.6 Element Type and Mesh Used in the Study ………………………………..31
3.6.1 Mathematical Details of Triangular Elements (CPE6) ………….. 32
3.6.2 Mathematical Details of Quadrilateral Elements (CPE8) ………... 34
3.6.3 Mathematical Details of 2-noded Beam Element …………………35
3.7 Material Properties ………………………………………………………….37
3.8 Loading and Boundary Conditions ………………………………………… 37
3.8.1 Loading ……………………………………………………………37
3.8.2 Boundary Conditions …………………………………………….. 38
3.9 Case Studies: Numerical Modeling of Unreinforced Slopes ………………. 39
3.9.1 Case 1: A Homogenous Slope with No Base ……………………..39
3.9.2 Case 2: A Homogenous Slope with Base ………………………... 42
3.10 Numerical Modeling of Reinforced Slopes ………………………………. 44
3.10.1 Influence of Soil Nail Length ……………………………………44
a) Benchmark with No Soil Base with One Nail ………………..44
b) Benchmark with Soil Base with One Nail ……………………45
3.10.2 Influence of Soil Nail Inclination ………………………………..46
3.10.3 Influence of Soil Nail Location ………………………………….47
3.10.4 Influence of Multiple Nails ……………………………………...47
CHAPTER 4: RESULTS and DISCUSSIONS ……………………………………. 49
4.1 Introduction ……………………………………………………………….. 49
4.2 Results for Unreinforced Slopes ………………………………………….. 49
4.2.1 Case 1: A Homogenous Slope with No Base …………………... 49
4.2.2 Case 2: A Homogenous Slope with Base ……………………….. 53
4.2.3 The Radius of failure circle compare with Limit Equilibrium
Method ………………………………………………………….56
4.3 Results for Reinforced Slopes ………………………………………………59
4.3.1 Influence of Soil Nail Length ……………………………………..60

v
 a) Benchmark with No Soil Base with One Nail ………………..60
b) Benchmark with Soil Base with One Nail ……………………63
4.3.2 Influence of Nail Inclination …………………………………….. 67
4.3.3 Influence of Nail Location ……………………………………….. 70
4.3.4 Influence of Number of Nails ……………………………………. 72
CHAPTER 5: SUMMARY AND CONCLUSIONS ………………………………... 74
5.1 Summary …………………………………………………………………… 74
5.2 Numerical Modeling of Unreinforced Slopes ……………………………… 75
5.2.1 Case 1: A Homogenous Slope with No Base ……………………..75
5.2.2 Case 2: A Homogenous Slope with Base …………………………76
5.3 Numerical Modeling of Reinforced Slopes …………………………………76
5.3.1 Influence of Soil Nail Length ……………………………………..76
a) Benchmark with No Soil Base with One Nail ……………….76
b) Benchmark with Soil Base with One Nail …………………..77
5.3.2 Influence of Nail Inclination ……………………………………...77
5.3.3 Influence of Nail Location ………………………………………..77
5.3.4 Influence of Number of Nails …………………………………….77
5.4 Conclusions …………………………………………………………………78
5.5 Recommendations …………………………………………………………..79
REFERENCES ……………………………………………………………………… 80
APPENDIX A: Influence of Soil Base on Factor of Safety ..………………….... 83

vi
 LIST OF TABLES

Table 2.1: Recommended Minimum Values of Factor of Safety (Duncan and

Wright 2005)…………………………………………………………….15

Table 2.2: Factor of Safety Criteria from U.S. Army Corps of Engineers’ Slope

Stability Manual …………………………………………………………16

Table 2.3: Reinforcing steel properties ASTM A615 (Grade 60& 75),

(FWHA, 1999)…………………………………………………………..21

Table 3.1: Soil properties …………………………………………………………..37

Table 3.2: Reinforcement properties ……………………………………………….37

Table 4.1: Computed Factor of safety for Case 1 ………………………………….50

Table 4.2: Computed Factor of safety results for Case 2 …………………………..54

Table 4.3: Computed Factor of safety for unreinforced slope and one-nail case …..62

Table 4.4: Factor of safety results for influence of number of nails on the factor

of safety ………………………………………………………………... 72

Table 5.1: Factor of safety results for homogeneous slope with no base (Case 1)….74

Table 5.2: Factor of safety results for homogeneous slope with base (Case 2)……. 75

Table A 1: Computed Factor of safety for Case 1 and Case 1 with a 5m

deep base added ………………………………………………………… 86

vii
 LIST OF FIGURES

Figure 1.1: Basic schematic diagram of a typical failure plain …………………1

Figure1.2: (a) Toe Failure, (b) Deep-Seated Failure, (c) Slope Failure, and
(d) Compound Failure (Duncan, 2005, Chen, 1995) …………………2
Figure1.3: Schematic diagram of a slope stability improvement by excavation.
(Duncan and Wright, 2005) …………………………………………..4
Figure1.4: Schematic diagram of a slope with berm …………………………….5

Figure1.5: Schematic diagram of a retaining wall (Das, 2004) ……………….....5

Figure1.6: Schematic diagram of a slope with vegetation ………………………6

Figure 1.7: Typical soil nailing detail, (modified after Broms and

Wong, 1991) …………………………………………………………7

Figure 2.1: Typical slope slip surface detail (modified after Duncan and

Wright, 2005) ………………………………………………………..17

Figure 2.2: Load transfer mechanism of soil nail structure, (Shiu and

Chang, 2004) …………………………………………………………19

Figure 2.3: A typical soil nail detail, (Shiu and Chang, 2004) ……………………19

Figure 2.4: Two commonly used soil nail section; Threaded Form (Williams
Form Engineering Corp., 2006) ……………………………………...20
Figure 2.5: Typical soil nail, and nail head detail (Broms and Wong, 1991) …….23

Figure 2.6: Typical punching shear of bearing plate (Broms and Wong, 1991) ….24

Figure 2.7: Typical facing pressure distribution on nail head detail (Broms
and Wong, 1991, Shong 2005) ………………………………………24
Figure 3.1: Nail head cross section used in the analysis (Abaqus, 2006) ………...30

Figure 3.2 a: Schematic detail of the connection of soil nail and the nail head…..30

Figure 3.2 b: Nail and nail head detail in FEA ……………………………………31

viii
Figure 3.3: Nail and nail head element mesh details ………………………………32

Figure 3.4: Quadratic triangle (Cook, et al. 2003) ………………………………..33

Figure 3.5: An 8-node quadrilateral element ………………………………...........35

Figure 3.6: 2-noded beam element ……………………………………………….35

Figure 3.7: Boundary conditions used in benchmark model with no

base soil ……………………………………………………………......38

Figure 3.8: Boundary conditions on benchmark model with base soil …………......38

Figure 3.9: Geometry of the benchmark model with no base model ……………....39

Figure 3.10: 6-node quadratic plane strain triangle elements with approximate
global size of 0.8 ……………………………………………………....40
Figure 3.11: 8-node biquadratic plane strain quadrilateral elements with
approximate global size of 0.8 ……………………………………......40
Figure 3.12: 6-node quadratic plane strain triangle elements with approximate
global size of 0.4 ………………………………………………….......41
Figure 3.13: 8-node biquadratic plane strain quadrilateral elements with
approximate global size of 0.4 …………………………………..........41
Figure 3.14: Geometry of the benchmark model with base soil ………………….....42

Figure 3.15: 6-node quadratic plane strain triangle elements (CPE6) ……………...43

Figure 3.16: 6-node quadratic plane strain triangle elements (CPE6) ……………...43

Figure 3.17: 6-node quadratic plane strain triangle elements (CPE6) ………...........44

Figure 3.18: Geometry of the benchmark with one soil nail. …………………….....45

Figure 3.19: Geometry of the benchmark with base with one soil nail. ………….....46

Figure 3.20: Geometry of the benchmark with base with inclined nail ………..........47

Figure 3.21: Geometry of the benchmark with base with different locations …........47

Figure 3.22: Geometry of the benchmark with base with multiple nails ……………48

ix
Figure 4.1 (a): Deformed mesh with CPE6 elements (Analysis 1) ……………….....50

Figure 4.1 (b): The equivalent plastic strain contour for Analysis 1………………....51

Figure 4.2 (a): Deformed mesh with CPE8 elements (Analysis 2) ……………….....51

Figure 4.2 (b): The equivalent plastic strain contour for Analysis 2………………... 51

Figure 4.3 (a): Deformed mesh with CPE6 elements (Analysis 3) ………………….52

Figure 4.3 (b): The equivalent plastic strain contour for Analysis 3………………....52

Figure 4.4 (a): Deformed mesh with CPE8 elements (Analysis 4) ………………….52

Figure 4.4 (b): The equivalent plastic strain contour for Analysis 4……………........53

Figure 4.5 (a): Deformed mesh with 728 elements (Analysis 1) …………………….54

Figure 4.5 (b): The equivalent plastic strain contour for Analysis 1…………………55

Figure 4.6 (a): Deformed mesh with 1164 elements (Analysis 2) …………………...55

Figure 4.6 (b): The equivalent plastic strain contour for Analysis 2…………………55

Figure 4.7 (a): Deformed mesh with 1397 elements (Analysis 3) …………………...56

Figure 4.7 (b): The equivalent plastic strain contour for Analysis 3…………………56

Figure 4.8: Limit equilibrium solution for Analysis 1 generated with

Geo-Studio, Slope/W ………………………………………………........57

Figure 4.9 (a): Displacement contours for Analysis 1 ……………………………....58

Figure 4.9 (b): The equivalent plastic strain contour for Analysis 1 ………………..59

Figure 4.10 (a): Deformed mesh for the case with a 10 m long nail ……………......60

Figure 4.10 (b): Displacement contours for the case with a 10 m long nail ………...61

Figure 4.11 (a): Deformed mesh for the case with a 15 m long nail ………………..61

Figure 4.11 (b): Displacement contours for the case with a 15 m long nail ………...61

Figure 4.12 (a): Deformed mesh for the case with a 20 m long nail ………………...62

x
Figure 4.12 (b): Displacement contours for the case with a 20 m long nail ………….62

Figure 4.13: Variation of factor of safety against length of the soil nail ……………...63

Figure 4.14: Schematic location of point “A” ………………………………………..64

Figure 4.15: Variation of displacement at point ‘A’ with the length of soil nail ……..64

Figure 4.16: The equivalent plastic strain contour for 4m long nail ……………….....65

Figure 4.17: The equivalent plastic strain contour for 6m long nail ……………….....66

Figure 4.18: The equivalent plastic strain contour for 8m long nail ……………….....66

Figure 4.19: The equivalent plastic strain contour for 15m long nail ………………...67

Figure 4.20: Variation of factor of safety with inclination angle of a soil nail …….....68

Figure 4.21: The equivalent plastic strain contours for nail inclination of

zero degrees ……………………………………………………………68

Figure 4.22: The equivalent plastic strain contours for nail inclination of 20 degrees

(counterclockwise) ……………………………………………………….69

Figure 4.23: The equivalent plastic strain contours for nail inclination of 45 degrees . 69

Figure 4.24: Variation of factor of safety with nail location …………………………..70

Figure 4.25: The equivalent plastic strain contours for the case in which nail was

located at the toe. ………………………………………………………...71

Figure 4.26: The equivalent plastic strain contours for one nail at 9m above from the

toe of slope ……………………………………………………………….71

Figure 4.27: The equivalent plastic strain contour for one nail at the toe of slope ……73

Figure 4.28: Displacement contour of three-nail case …………………………………73

Figure A 1: Geometry of the benchmark model with no base model (Case 1) ………. 83

xi
Figure A 2: Geometry of the benchmark model with a 5m deep base added ………… 83

Figure A 3: Deformed mesh for Case 1 (from Analysis 1) …………………………… 84

Figure A 4: Deformed mesh for Case 1 with base ……………………………………. 84

Figure A 5: Displacement contours …………………………………………………… 85

Figure A 6: The equivalent plastic strain contours ……………………………………. 85

xii
 CHAPTER 1
INTRODUCTION

1.1 Introduction

Analysis of the stability of complex soil slopes is an important and challenging

aspect of geotechnical engineering. Over the past 70 years, extensive engineering and
research studies have been conducted to develop conventional methods for analysis of
slope stability (Duncan and Wright, 2005). Although several advances have been made
over the past decades, slope stability analysis still remains a challenge in geotechnical
engineering (Duncan and Wright, 2005).

Depending on the geometric details and soil properties, slope failures occur along
a failure surface in the soil mass as shown in Figure 1.1. Slope stability analyses by
either limit equilibrium method or finite element method using the shear strength
reduction (SSR) technique can be used to assess the original factor of safety and the
improvement after strengthening (Dawson et al, 1999; Griffith and Lane, 1999).

Typical failure plane

Figure 1.1: Basic schematic diagram of a typical failure plane

1
1.2 Types of Slope Failures and Instability Mechanism

A common mode of slope failure is a rotational slip along an almost circular

failure surface (Duncan, 2005). Brief descriptions of the different types of slope failures
are given below:

• Toe Failure, in which failure occurs along the surface that passes through the toe
[Figure1.2 (a)]
• Base Failure, in which the failure surface passes below the toe [Figure1.2(b)]
• Slope Failure, in which the failure occurs along a surface that intersects the slope
above the toe [Figure1.2 (c)]
• Compound Failure, in which the failure is a combination of the rotational slips
and the translational (infinite) slip. Generally occurs when a hard stratum exists at
a quite large distance from the slope surface. [Figure1.2(d)]

(a) (b)

(c) (d)

Figure1.2: (a) Toe Failure, (b) Deep-Seated Failure, (c) Slope Failure, and
(d) Compound Failure (Duncan, 2005; Chen, 1995).

2
 It is important to understand the causes of instability in slopes for purposes of
designing and constructing new slopes, and for the purposes of repairing failed slopes.
For stability of a slope, the shear strength of the soil must be greater than the shear stress
requirement for equilibrium. The instability condition can be reached through two
mechanisms (Duncan and Wright, 2005):

• A decrease in the shear strength: The loss of the shear strength may occur due
to an increase in moisture content, pore water pressure, shock or cyclic loads,
weathering, etc.
• An increase in the shear stress: The stresses may increase due to weight of
water causing saturation of soils, surcharge loads, seepage pressure, etc.

In most cases, several causes exist simultaneously. According to Sowers (1979),

it is usually not possible to identify the cause that acted alone and resulted in instability.
Attempting to identify which one finally produced the failure is not only intricate, but
also technically inaccurate (Duncan and Wright, 2005). Hence, in designing and
constructing new slopes, it is important to consider potential changes in properties and
conditions that may affect the structure during its lifespan so that it will remain stable
regardless of these changes.

3
1.3 Conventional Techniques to Improve the Stability of Slopes

There are many methods available to improve the slope stability (Duncan and
Wright, 2005). Brief descriptions of the different methods are given below:

• Unloading, which reduces the weight of the mass tending to slide.

Excavate top

Excavate bench

Flatten slope

Figure1.3: Schematic diagram of a slope stability improvement by excavation

(Duncan and Wright, 2005)

4
• Placing a berm at the toe to increase the resisting forces. It is especially useful
when base failure is anticipated (Figure1.4).

Berm

Figure1.4: Schematic diagram of a slope with berm

• Drainage, which helps reducing the seepage forces that may increase the stability,
the zone of subsurface water is lowered and infiltration of the surface water is
prevented.
• Reinforcement and retaining walls which help increase stability by providing
lateral support (Figure1.5).

Figure1.5: Schematic diagram of a retaining wall (Das, 2004)

5
 • Vegetation, which reduces the effects of runoff on the slope by intercepting
precipitation. It can only provide surficial stabilization within the root zone
(Figure1.6).

Vegetation

Root zone

Figure1.6: Schematic diagram of a slope with vegetation

• Soil hardening, which can be provided by using grouting and injecting of cement
or other compounds into specific zones to increase the stability.
• Densification by explosives, vibroflotation, or terra probe, which helps increase
the shear strength in cohesionless soils.

Reinforcements can be used in many different applications in geotechnical

engineering such as in embankments, cut slopes, soil slope reinforcement, and retaining
walls. The use of soil nailing is becoming a rapidly growing technology in soil slope
stability techniques.

Soil nails are either installed in drilled bore holes or secured with grout, or they are
driven into the ground. The soil nails are generally attached to concrete facing located at
the surface of the structure (Broms and Wong, 1991) (Figure 1.7).

6
 Soil nail before
shotcrete

Steel rebar Bore hole filled

(nail) with grout

Figure 1.7: Typical soil nailing detail, (modified after Broms and Wong, 1991)

Soil nailing is one of the in-situ methods to reinforce ground by installing closely
spaced reinforcing elements, usually steel bars, to increase the overall shear strength of
the soil and reduce potential displacements. The method is an effective process for
retention of deep excavations, stabilization and remediation of slopes when properly
engineered (Ann et al, 2004). The structural reinforcing elements provide resisting
tensile forces, shear stresses and bending moments through the friction mobilized at the
interfaces of nail and in-situ soil. It can improve soil performance, increase the safety
factor, and reduce the construction cost for a project (Goldstein, 2001).

7
1.4 Factor of Safety

The factor of safety (Fs) is very important when performing slope stability
analyses. The most widely used definition of factor of safety for a slope is the ratio of
shear strength of the soil to shear stress required for equilibrium (Duncan and Wright,
2005). It can be determined from a limit equilibrium analysis using factored strength
parameters (Griffiths and Lane, 1999). For a slope analysis a unique factor of safety can
be determined using conventional methods based on limit equilibrium methods (Ann et
al, 2004).

The classical approach for factor of safety of the slope is determined by

considering a range of failure surfaces with different centers and radii to find the slip
circle that gives the minimum factor of safety. However, these methods cannot be easily
used for complex geometries and reinforced slopes.

1.5 Problem Statement

Evaluation of the stability of soil slopes using the finite element method has
gained popularity in recent years. The slope stability analyses are conventionally assessed
using Limit Equilibrium Methods (LEM) (Ann et al, 2004). Limit Equilibrium Methods
have limitations in dealing with this type of strain and compatibility problems in
reinforced soil slopes.

In this research, three different types of slopes were analyzed by using the finite
element method.

1. A benchmark problem
2. An embankment with no reinforcement
3. An embankment with soil nail(s)

8
 The first model was created to study the effects of meshing on slope factor of
safety. It was compared with conventional methods and previous research findings. The
second model and the third model were created to better understand the stability
improvements by using soil nails, and their influence on failure plane and the factor of
safety. The second model was also compared with conventional methods and previous
research findings. Soil-nailed embankment included the following aspects: Influence of
nail inclination, nail location, and nail length.

The finite element method is and non-associative Mohr-Coulomb plasticity were

used in all models throughout this paper. Numerical simulations were performed to
investigate states of failure and investigate the location of the slip surface due to strength
parameters and existing conditions. Using the numerical analysis, it was possible to
obtain an accurate factor of safety of slopes analyzed in previous research studies by
using conventional methods such as, Bishop’s Modified Method (Bishop, 1955), Janbu’s
Modified Method (Janbu, 1968), Morgenstern and Price’s Method (Morgenstern and
Price, 1965), and Spencer’s Method (Spencer, 1967). In addition to these methods the
General Limit Equilibrium (GLE) is mentioned. These methods generally differ from one
to another in the equations of static equilibrium and the relationship between the
interslice and the shear forces (Shiu et al., 2006). A difficulty with all the conventional
methods that they are based on assumption of the shape or the location of failure surface
the (Griffiths and Lane, 1999).

The computer software package GeoStudio 2004-SLOPE/W developed by GEO-

SLOPE was used in determining the factor of safety values are obtained from the limit
equilibrium methods that were mentioned above.

9
1.6 Scope of Work

A literature review was conducted to investigate deformation-based slope stability

analyses, effects of meshing, and to investigate the stress distribution in the nails and
resulting improvement in stability. A well known finite element package ‘ABAQUS’ was
used to calculate the factor of safety for both reinforced and unreinforced slopes. The
factors of safety obtained from different theories and previous research studies were
compared with the numerical results to assess how close or far the slopes are from failure.
Also, the results obtained from the finite element analyses were compared with results
obtained in previous studies (Griffiths and Lane, 1999). Factor of safety was determined
by using the shear strength reduction (SSR) technique (Griffiths and Lane, 1999; Dawson
et al, 1999). Two-dimensional models were used in all cases presented in this paper.

1.7 Research Objectives

The primary objectives of this study were:

• Review existing literature of slope stability analysis and procedures

• Select an appropriate element type to create the models
• Perform finite element analysis for each model
• Study the influence of density of meshing and reinforcement on factor of safety
values as well as on the failure mechanism
• Study the effects of soil nails on slope factor of safety and failure zone
• Compare the findings from finite element analysis with previous research studies.

10
1.8 Previous Studies

In one particular study (Griffiths and Lane, 1999), slope stability analysis by finite
element analysis was performed on several examples. This study compared the finite
element results with traditional methods of slope stability analysis. The slope stability
examples were performed on both slopes with no foundations and slopes with
foundations using various material properties. The study concluded that the finite
element method in conjunction with an elastic-perfectly plastic (Mohr-Coulomb) stress-
strain method was reliable and robust assessing the factor of safety of slopes.

In another study (Ann et al, 2004), a finite element analysis of a soil nailed slope
was investigated. Reinforced slope stability analysis by using finite element method and
limit equilibrium method were compared. Sensors and strain gages were located along
selected soil nails. The mobilization of soil nail forces at various stages of excavation and
the lateral displacement of the nailed slope due to stress–relief was monitored. The
results show that both FEM and LEM methods provided similar factor of safety values
and failure mechanism pattern for 30 degree slope face. The factor of safety values were
dissimilar for a steeper slope face of 75 degrees.

Another study examined the load transfer mechanism of soil nails and nail heads,
and the effect of nail inclination on slope stability using numerical analysis (Shiu and
Chang, 2004). Numerical simulations were performed on Fast Lagrangian Analysis of
Continua (FLAC), and Mohr-Coulomb model was used for the soil. Two dimensional
plane strain analysis was used in numerical analyses. The study examined nailed-slopes
with no nail head and with nail heads and it was concluded that the factor of safety
increased when nail heads were introduced to soil nails. Also, slopes with no
reinforcement and with reinforcement were analyzed. The factor of safety of
unreinforced slope increased from 1.0 to 1.2 when soil nails were introduced to the
analyses.

11
 CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

In geotechnical engineering, slope stability problems are affected by shear

strength of the soil. Shear strength of soils is the result of the resistance to mass
deformation formed from movement of particles, particle interlocking, physical bonds,
and chemical bonding across particles (Terzaghi, 1996). The shear strength of soil is
usually expressed in terms of two soil parameters: angle of friction (ø) and cohesion (c).

The finite element analysis can be used to compute stresses and displacements in
a soil mass subjected to external loads. However, the stability of a slope cannot be
determined directly from the finite element analysis. Only, the computed stresses in a
slope can be used to obtain a factor safety. In this sense, finite element analysis is a
complex method for computing factor of safety of slope stability problems (USACE,
2003). In the present research paper, the Shear Strength Reduction (SSR) technique was
used to compute the factor of safety based on the finite element analysis.

2.2 Shear Failure

A shear failure involves sliding of a portion of an embankment, or an

embankment and its foundation, relative to the adjacent stationary mass. A shear failure
is usually considered to occur along a discrete surface as assumed in stability analyses,
although the shear movements may in fact occur across a zone of appreciable thickness.
Failure surfaces are frequently approximately circular in shape. For soils, failure occurs
at 15% to 20% of strain (Holtz and Kovacs, 1981). Failure of a soil element at a certain
location does not mean failure of the system. However, it could mean a reduction of the
factor of safety.

12
2.3 Mohr–Coulomb Model Description

There are different criteria that can define failure. Some of these criteria are
based on strain level, while most are based on the shear stress in comparison to the shear
strength. The Mohr-Coulomb failure criterion is the most commonly used one in soil
mechanics. The Mohr-Coulomb failure model was used in this study. It is defined as
(Das, 2002):
τ f = c'+σ ' f tan φ ' …………………………………………… 2.1

Where τ f shear strength at failure, c' is effective cohesion, σ ' f is effective

stress at failure, and φ ' is the effective angle of friction. The Mohr-Coulomb model is a
perfectly elastic-plastic model. It has been widely used in geotechnical applications
(Hammah et al, 2004).

2.4 The Shear Strength Reduction (SSR) Technique

Shear Strength Reduction method is a new technique in finite element method to

obtain factor of safety of earth slopes. The finite element method was first applied to
geotechnical engineering in 1966 (Rocscience, 2004). In the mid 1970s, slope stability
analysis started appearing in the literature. According to the SSR method, soil shear
strength is reduced to bring a slope to verge of failure (Duncan and Wright, 1996). In the
finite element method, such a state is detected by the inability to reach equilibrium. In
the SSR technique, it is assumed that the slope materials have elasto-plastic behavior.
The material strength is reduced until failure occurs.

13
 In this technique, the Mohr-Coulomb material shear strength is reduced by a
factor F, which is also called the “factor of safety”. The definition is expressed as:

τ c' tanφ '

= + ………………………………………2.2
F F F

The equation can be written as

τ
= c* + tanφ * ………..…………………………2.3
F

Where
c'
c* = ……….……………………2.4 (a)
F

⎛ tanφ ' ⎞
φ * = arctan⎜ ⎟ ……….……………………2.4 (b)
⎝ F ⎠

For Mohr-Coulomb materials, the steps for systematically searching for the
critical factor of safety value, F, which brings a previously stable slope to the verge of
failure, are given below (Rocscience. 2004):

Step 1: Develop an FE model of a slope, using the deformation and strength

properties established for the slope materials. Compute and record the maximum
total deformation in the slope by using the finite element method.

Step 2: Increase the value of F and calculate factored Mohr-Coulomb material

parameters as described above. Enter the new strength properties into the slope
model and re-compute deformation. Record the maximum total deformation.

14
 Step 3: Repeat Step 2, using systematic increments of F, until the FE model does
not converge to a solution. In other words, continue to reduce material strength
until the slope fails. The critical F value beyond which failure occurs will be the
slope factor of safety, based on the finite element method.

For a slope that is initially unstable, factor of safety values in steps 2 and 3 must
be reduced until the finite element model converges to a solution (Rocscience, 2004).

2.5 Determination of Factor of Safety (FOS)

The most widely used definition of factor of safety for a slope is the ratio of shear
strength of the soil to shear stress required for equilibrium. Shear strength is often the
largest uncertainty in slope stability analyses. A value of F=1.0 indicates that a slope is
on the boundary between stability and instability. If all the factors are computed
precisely, even a value of 1.01 would be acceptable. However, the computed values of
FOS are not precise, due to uncertainty of variables. Therefore, the factor of safety
should be larger to be on the safe side (Duncan and Wright, 2005).

The value of factor of safety should be matched to uncertainty of analysis

conditions. Recommended minimum values of factor of safety are shown in Table 2.1
(Duncan and Wright, 2005).

Table 2.1: Recommended Minimum Values of Factor of Safety (Duncan and

Wright 2005)

Cost and consequences of slope failure Uncertainity of analysis conditions

Small Large
Cost of repair comparable to incremental 1.25 1.5
cost to more conservatively designed slope

Cost of repair much greater than 1.5 2.0 or greater

incremental cost to construct more
conservatively designed slope

15
 The recommended values of factor of safety shown in Table 1.2 from the U.S.
Army Corps of Engineers’ slope stability manual is based on experience.

Table 2.2: Factor of Safety Criteria from U.S. Army Corps of Engineers’ Slope
Stability Manual

Types of Slopes Required factors of safety

For end of For long-term For rapid
construction steady seepage drawdown
Cost of repair comparable to
incremental cost to more 1.3 1.5 1.0 -1.2
conservatively designed slope

Another approach of factor of safety of slopes refers to the ratio of resisting

moment to overturning moment on circular slip surfaces (Duncan, 2005).

Referring to the slope and circular slip surface shown in Figure 2.1, factor of
safety, F can be written as:

available resisting moment

F= ……………………………… 2.4
actual driving moment

Available resisting moment,

clr
Mr = …………………………………………………….. 2.5
F

Where, c is cohesion, l is the length of the circular arc, and r is the radius.

Actual driving (overturning moment),

Md= Wχ ………………………………............................. 2.6

Where, W is the weight of the soil mass and χ is the moment arm.

clr
F= ………………………………………………… 2.7
Wχ

16
 Circle
center χ

θ r

W Circular failure
surface, length l
S

Figure 2.1: Typical slope slip surface detail (modified after Duncan and Wright,
2005)

When a soil mass is reinforced with reinforcing elements such as soil nails, the
shear failure is much more complex. Different factors of safety can be used for
reinforcement and soil strength (Duncan and Wright, 1996). This requires a more detailed
treatment in the finite element analysis.

2.6 Slope Stability with Soil Nails

Soil nailing is a slope stabilization method that uses an array of nails inserted into
the ground (Goldstein, 2001). Soil nailing can be used for both temporary and permanent
slopes (Chance, 1999). It also can be used to support slopes or to repair slipping
embankments and gravity walls. Soil nailing is commonly used in soft rocks and stiff
clays. Soil nails tend to work best in areas where there is a high content of clay material
in the soil (KUTC, 2006).
2.6.1 Historical Background of Soil Nailing

17
 The soil nailing technique was developed in late 1960’s and has been popular
throughout the world (Chance, 1999). Soil nailing originated in Europe in the early 1960s
with the introduction of the New Austrian Tunneling Method (NATM) (Karakus and
Fowell, 2004). Rock bolts and shotcrete have been used to support the tunnels, and the
soil nailing emerged as a result of the experience from rock bolts. This technology was
reportedly first used to support retaining walls in France in 1961 (Chance, 1999). The use
of grouted soil nails and driven soil nails, which consist of solid steel bars and steel iron,
continued to grow in the 1960s in France and Germany. It has been reported that soil
nails were used to support a wall in grouted sand in France in 1972 (Chance, 1999). This
60-foot wall was sitting on a 21 degree slope to support an excavation of a railroad track
(Banerjee et al., 1998).

The first application of soil nailing in North America seems to be a temporary

retaining wall in Vancouver, BC, in the late 1960s (Chance, 1999). A soil nailing
construction project was for an excavation of a hospital foundation in Portland, OR was
reported in 1976 (Chance, 1999). According to the reports from project managers in
Portland, the construction time was reduced by up to 50% and yielded a cost savings of
15% over conventional tieback construction by using soil nailing (Goldstein, 2001).

2.6.2 Reinforcing Mechanism in Soil Nailing

The stability of a soil mass reinforced with soil nails improve due to two
mechanisms: the transfer of tensile forces generated in the nails due to frictional
interaction between the ground and the nail, and the development of shear stress and
bending stiffness in the nails as a result of deformation of soil mass (Figure 2.2). In
addition to these mechanisms, the soil-structure interaction between the facing and the
soil helps to restrain displacement and produce nail head load at the connection between
the nail and facing to develop the force along the nails (Byrne et al, 1996).

18
Figure 2.2: Load transfer mechanism of soil nail structure, (Shiu and Chang,
2004)

Figure 2.3: A typical soil nail detail, (Shiu and Chang, 2004)

19
2.6.3 Soil Nail Installation

Soil nails are installed in a pattern designed to ensure both internal and external
stability of a wall. A fairly large number of nails are placed so they can resist the tensile,
compressive, and shear stresses within the wall and transfer them into the ground (Figure
2.3).

Steel reinforcing bars used for soil nails are commonly threaded and may be
either solid or hollow. Bars generally have a nominal tensile strength of 420 MPa [60
kips per square inch (ksi) or Grade 60] or 520 MPa (75 ksi or Grade 75). The common
U.S. practice of soil nailing involves the use of solid steel bars of 420 or 520 MPa (Grade
60 or 75) (FWHA, 1999).

Threaded bars are generally used in practice. The two commonly used threaded
soil nail sections shown in Figure 2.4. Reinforcing steel properties for Grade 60 and 75
(ASTM A615), can be seen in Table 2.1.

Figure 2.4: Two commonly used soil nail section; Threaded Form (Williams
Form Engineering Corp., 2006)

20
 Table 2.1: Reinforcing steel properties ASTM A615 (Grade 60& 75), (FWHA,
1999).

Grade Threadbar Yield Cross Yield Nominal Max.Threadbar

Designation Stress Sectional Strength Weight Diameter
Area
in- mm ksi Mpa in² mm² kips kN lbs/lf kg/m in mm
lb
60 #6 19 60 414 0.44 284 26.4 118 1.50 2.24 0.86 21.8
60 #7 22 60 414 0.60 387 36.0 160 2.04 3.04 0.99 25.1
60 #8 25 60 414 0.79 510 47.4 211 2.67 3.98 1.12 28.4
75 #6 19 75 517 0.44 284 33.0 147 1.50 2.24 0.86 21.8
75 #7 22 75 517 0.60 387 45.0 200 2.04 3.04 0.99 25.1
75 #8 25 75 517 0.79 510 59.3 264 2.67 3.98 1.12 28.4
75 #9 29 75 517 1.00 645 75.0 334 3.40 5.06 1.26 32.0
75 #10 32 75 517 1.27 819 95.3 424 4.30 6.41 1.43 36.3
75 #11 36 75 517 1.56 1006 117.0 520 5.31 7.91 1.61 40.9
75 #14 43 75 517 2.25 1452 168.8 751 7.65 11.39 1.86 47.2

There are five types of soil nails: Driven nails, grouted nails, corrosion-protected
nails, jet grouted nails, and launched nails (Goldstein, 2001).

Driven nails are generally small-diameter nails (15-46 mm) with a relatively limited
length (to about 20 m) made of mild steel (about 50 ksi) that are closely spaced in the
wall (two to four nails per square meter). Nails with an axial channel can be used to
permit the addition of grout sealing. This installation method is the quickest (four to six
per hour) and most economical to establish with a pneumatic or hydraulic hammer.

Grouted nails are generally steel bars, with diameters ranging from 15 to 46 mm, stronger
than driven nails (about 60 ksi). Grouted nails are inserted into boreholes of 10-15 cm
and then cement-grouted by gravity and under low pressure. Ribbed bars are also used to
increase soil adhesion.

21
 Corrosion-protected nails are generally used for aggressive soils as well as for
permanent structures. Corrosion control is necessary for long-term temporary and/or
permanent works and can be prevented by using one of these methods: epoxy coatings,
polyethylene sheathing, or chatodic protection.

Jet-grouted nails are composite inclusions made of grouted soil and a central steel
rod that can be as thick as 30 to 40 cm thick. During installation, nails are installed using
a high-frequency vibropercussion hammer, and cement grouting is injected. The jet-
grouted installation technique has been shown to increase the pullout resistance of the
composite, and the nails are corrosion-resistant.

Launched nails are between 25 and 38 mm in diameter and up to 6 m or longer

are fired directly into the soil, using a compressed-air launcher. During installation, the
ground around the nail is displaced and compressed. This installation technique involves
the least site disturbance.

2.6.4 Nail Heads

The strength of nail head is one of the parameters used in design of soil-nailing
practice (Shiu and Chang, 2004). The strength of the nail head determines the stability of
the nailed structure against failure involving the facing element. A typical facing system
is shown in Figure 2.5. Design of the nail head is out of scope of this study, but it should
be noted that the nail head plays an important role in the design of the soil-nail system.

22
 Shotcrete
Ungrouted area
packed with soil
Steel
Plate

Steel
Rebar

Bore hole filled

with grout

Figure 2.5: Typical soil nail, and nail head detail (Broms and Wong, 1991)

Strength of nail head is primarily governed by the flexure and/or punching shear
of the facings, and nail head connection (Shong, 2005).

The flexural strength of the proposed facing system for a soil nail wall must be
analyzed to assure that the loads generated by the non-uniform earth pressure between the
nails can be resisted without flexural failure of the facing (Chance, 1999). A typical
facing system is shown in Figure 2.6. The structural capacity of the facing can, therefore,
be determined using standard reinforced concrete design procedures for singly-
reinforced, rectangular concrete beams (Shong, 2005).

23
 Punching
shear surface

Figure 2.6: Typical punching shear of bearing plate (Broms and Wong, 1991)

Pressure built-up
at nail head

Figure 2.7: Typical facing pressure distribution on nail head detail (Broms and
Wong, 1991, Shong 2005)

Punching shear failure of the connection of a soil nail system is shown in Figure
2.7, and involves punching a cone of shotcrete centered about the nail head through the
facing (Chance, 1999). There are two components of the resistance of the system to
punching shear; the resistance provided by the facing (shotcrete and reinforcing steel);
and the resistance provided by the soil behind the facing. The analysis procedure
presented herein ignores the contribution of the soil in determining the punching shear

24
strength because the soil at the face of the wall is generally disturbed by the installation
of the anchor and may provide little or no bearing resistance (Shong, 2005). It also
assumes that the square bearing plate may be represented by a circular plate with a
diameter equal to the width of the plate and that welded wire mesh steel reinforcement
does not provide any shear capacity reinforcement (Chance, 1999).

25
 CHAPTER 3
NUMERICAL METHODOLOGY

3.1 Introduction

Finite element method (FEM) is a powerful numerical tool for solving many
engineering problems dealing with mechanical behavior of solids. With recent
advancements in computer technology, the finite element method has become a popular
alternative to traditional methods in geotechnical engineering. In this research, a finite
element method was used to numerically obtain the factor of safety for both reinforced
and unreinforced slopes. The factors of safety obtained from different theories and
previous research work are compared with the numerical results from the present study to
evaluate soil slope failures. Factor of safety was determined by using the shear strength
reduction (SSR) technique, which was described in Chapter 2.

3.2 Finite Element Method (FEM)

Finite element method (FEM), also called finite element analysis (FEA), is a
numerical method for solution of complex problems. Individual finite elements can be
visualized as small pieces of a structure. The first step of a finite element analysis is to
divide the actual geometry of a structure using a collection of discrete portions called
“finite elements”. Elements are connected at points called nodes. The collection of the
nodes and finite elements is called the mesh. The number and the type of the elements
need to be carefully chosen to effectively approximate the variables over the region of
interest. The governing equations for each finite element are determined and assembled
to analyze the behavior of the solid body (Cook et al. 2003), which is subjected to
external loads and boundary conditions.

26
 The governing equation for the discritized domain can be written as:

[K ] {r}= {R} ……………………………………………… 3.1

where,

[K ] = global stiffness matrix

{r} = global displacement vector
{R} = global load vector

In the present study, the constitutive model used for soils was the Mohr-Coulomb
criteria with non-associative plastic flow (Griffiths and Lane, 1999). Therefore, the
resulting [K] matrix is non-symmetric (Abaqus, 2006), and the equations solved by using
a non-symmetric solver (Abaqus, 2006).

3.3 Geometry

A number of soil problems including reinforced soil slopes were solved in this
research. The novelty of this research is the analysis of reinforced slopes. However, as a
first step, a number of previously solved soil slopes were re-analyzed to calibrate the
present computational model. Details of some of these problems can be found elsewhere
in the literature (Griffiths and Lane, 1999; Ann et al, 2004; Shiu and Chang, 2004). Two
different soil slope models are presented in this paper as shown in Figure 3.5 and Figure
3.6. These models are: (a) a benchmark problem with no base, and (b) a benchmark with
base (foundation soil).

The analysis of a reinforced soil slope includes three different parts: (a) soil, (b)
soil nail, and (c) nail head. These parts are described below.

27
3.4 Soil Model

The soil is considered as a non-linear elasto-plastic material in this research. In

general, soils do not behave like a linear-elastic material (Yu, 2000). The analysis for the
soil is based upon the Mohr-Coulomb failure criterion.

The Mohr-Coulomb criterion can be expressed as (Das, 2002):

τ f = c + σ tan φ ………………………………………… 3.2

where:
c = cohesion
φ = angle of internal friction
σ = normal stress
τf = shear stress

In this paper, the soil was modeled as a homogeneous, two-dimensional, plane

strain medium. It was assumed that the soil properties do not depend on temperature. In
description of the material properties of soil, non-associated Mohr-Coulomb plasticity
model is used. The soil model included six parameters:

φ ' : Angle of friction (degree)

c' : Cohesion
ψ: Dilation angle (degree)
ν: Poisson’s ratio
E: Young’s modulus (kN/m2 , psi)
γ: Unit weight (kN/m3 , pcf)

The relationship between the dilation angle ψ, and the angle of friction φ ' ,
determines whether the soil dilates or compacts upon plastic deformation (Fenton, 1990).

28
If the ratio of dilation angle to the angle of friction (ψ/ φ ' >1.0) exceeds 1.0, then only
compaction occurs. Otherwise initial compaction is followed by dilation (Fenton, 1990).
Dilation is a measure of how much volume change takes place when the material
undergoes shearing (Griffiths and Lane, 1999). For a Mohr-Coulomb material, dilation is
an angle that generally varies between zero and the friction angle. If ψ= φ ' , the plastic
flow occurs according to “associative flow rule”. If, ψ=0, then the plasticity corresponds
to a “non-associated flow rule” (Griffiths and Lane, 1999). Previous studies on soil
dilatancy were mostly concentrated on the theoretical analysis of the dilation angle, and
its influence on soil strength (Chen et al, 2003). Very few actual test data of dilation
angles were reported (Chen et al, 2003). In this study, the volume change in soil during
the failure was not considered. Therefore, dilation angle, ψ, was taken as zero.

Poisson’s ratio, ν, and Young’s ratio, E, are the elastic parameters of the soil.
Although, these parameters have a significant influence on the computed deformations
prior to failure, they have very little influence on the predictions of factor of safety of
slopes (Griffiths and Lane, 1999). In this study, constant values for soil modeling were
used for these two parameters, as given below:

E = 105 kN/m2
ν = 0.3

The most important parameters in the finite element analysis of slope stability
analysis are: angle of friction φ ' , cohesion c' , and unit weight γ, and geometry of the
model (Griffiths and Lane, 1999). Different values of these parameters were used in the
problems solved in this research. Details of the problems solved in this research can be
found in the next chapter.

29
3.5 Nail and Nail Head Model

Nail and nail head geometric parameters were taken published literature on a
“reinforced slope model” (Abaqus, 2006). Although, designing nails and nail heads are
not in the main focus of this research, selection of appropriate parameters is an important
task. Pre-stressed nails can be used in soil-nailed slope models in order to evaluate their
effects on the failure plane and the factor of safety of slopes (Abaqus, 2006). Based on
published data (Abaqus, 2006), it was assumed that the nails and the nail heads have a
3
density of 7,800 kg/m , and an elastic modulus of 200 GPa. The nails are considered as
embedded, which means they are fully bonded to the surrounding soil. Tips of nails were
coupled to the upper surface of the nail heads (Figure 3.2 (b)). Nail heads were tied to
the upper face of the soil slope (Figures 3.2 (a) and 3.2 (b). Nail length was varied to
investigate the influence of nail length on slope stability. The length was chosen as 20
meters in all other cases for consistency among each model used in this study. Nail head
parameters were kept the same in all cases.

1.6 m
0.005 m

Figure 3.1: Nail head cross section used in the analysis (Abaqus, 2006)

Nail head Nail tip

Nail

Figure 3.2 (a): Schematic detail of the connection of soil nail and the nail head

30
 Nail tip is coupled
to nail head upper
surface

Nail

Nail head

Figure 3.2 (b): Nail and nail head detail in FEA

3.6 Element Type and Mesh Used in the Study

Element type and mesh properties did not differ for each model in this study. The
soil body (slope) was modeled by using 6-noded quadratic triangular plane strain (CPE6)
elements (Figures 3.10, and 3.12). Moreover, 8-noded quadrilateral plane strain (CPE8)
elements (Figures 3.11, and 3.13) were also used in the benchmark model with no base to
investigate the effects of element type on computed factor of safety. Soil nails were
modeled by using 2-noded linear beam (B21) in this study. Nail head was modeled by
using 8-noded biquadratic plane strain (CPE8) elements. Typical finite element meshes
for nail and nail head used in the study are shown in Figure 3.3.

31
 8-noded biquadratic
plane strain elements
(CPE8) 2-noded linear
beam elements
(B21)

Nail head
Nail

Figure 3.3: Nail and nail head element mesh details

3.6.1 Mathematical Details of Triangular Elements (CPE6)

A quadratic triangle has side nodes in addition to corner nodes as shown in Figure
3.4. For stress analysis, nodal unknowns (degree of freedom) are u i and ν i at each node
i=1, 2,…, 6, can be expressed as (Cook, et al. 2003):

u = N 1U 1 + N 2U 2 + ... + N 6U 6 ..…………………. 3.2

ν = N 1V1 + N 2V2 + ... + N 6V6 ……………………. 3.3

where,
u, ν represent the nodal displacements in the x- and y- directions
Ni, represents interpolation function.

32
 ν3

u3
u6 3
ν5
ν6
6
5 u5
u1
y,ν 1
4
ν1 u4
2
u2
ν4
x,u
ν2

Figure 3.4: Quadratic triangle (Cook, et al. 2003)

Element strains can be expressed in terms of displacements as (Cook, et al. 2003):

δu
εx = ……..…………………. 3.4 a
δx

δv
εy = ……..…………………. 3.4 b
δy

δu δv
γ xy = + …………………. 3.4 c
δy δx

33
3.6.2 Mathematical Details of Quadrilateral Elements (CPE8)

A quadratic rectangle can be achieved by adding additional side notes to the linear
rectangle. There are 8 nodes in quadrilateral elements as shown in Figure 3.5. The
displacement field of the element can be expressed as (Cook, et al. 2003):

u = N 1U 1 + N 2U 2 + ... + N 8U 8 ..…………………. 3.5

ν = N 1V1 + N 2V2 + ... + N 8V8 …………………….. 3.6

where,
u, ν represent the nodal displacements in the x- and y- directions
Ni, represents interpolation function.

Strains in the element can be expressed as:

∂N 1 ∂N 2 ∂N 8
εx = U1 + U 2 + ... + U 8 ….…………………………….. 3.7 a
∂N x ∂N x ∂N x

∂N 1 ∂N 2 ∂N 8
εy = V1 + V2 + ... + V8 …………………………………… 3.7 b
∂N y ∂N y ∂N y

∂N 1 ∂N 2 ∂N 8 ∂N 1 ∂N 2 ∂N 8
γ xy = U1 + U 2 + ... + U8 + V1 + V2 + ... + V8 .. 3.7 c
∂N y ∂N y ∂N y ∂N x ∂N x ∂N x

34
 Figure 3.5: An 8-node quadrilateral element

3.6.3 Mathematical Details of 2-noded Beam Element

A 2-noded beam element has a node at each end and each node has two degree of
freedom as shown in Figure 3.6. The stiffness matrix of a 2D beam element is 4 by 4, and
the load vector contains moments and forces.

y¸ν y¸ν

θ z1 θ z2 M1 M2
E, I z E, I z
1 χ 1 χ
2 2
ν1 ν2 P1 P2
L L

(a) (b)

Figure 3.6: 2-noded beam element

35
The complete 2-noded beam element stiffness matrix can be written as (Cook, et al.
2003):

⎡ 12 EI z 6 EI z − 12 EI z 6 EI z ⎤
⎢ L3 L2 L3 L2 ⎥
⎢ ⎥
⎢ ⎥
⎢ 6 EI z 4 EI z − 6 EI z 2 EI z ⎥
⎢ L2 L L2 L ⎥⎥
⎢
[K] = ⎢ ⎥ …………………… 3.8
⎢ − 12 EI z − 6 EI z 12 EI z − 6 EI z ⎥
⎢ 3 ⎥
⎢ L L2 L3 L2 ⎥
⎢ ⎥
⎢ 6 EI z 2 EI z − 6 EI z 4 EI z ⎥
⎢ ⎥
⎣ L2 L L2 L ⎦

Where,

E = elastic modulus
I z = moment of inertia of the beam cross-sectional area about a centroidal
axis parallel to the z axis.
ν1, ν2 = lateral translations
θ z1 , θ z 2 = rotations of the beam element

The governing equation can be expressed as:

[K]{q}={Q} …………………………………………………….. 3.9

{q}= {ν1, θ z1 , ν2, θ z 2 }T ….……………………………………… 3.10

{Q}= {P1, M1, P2, M2} …………………………………………. 3.11

where,
[K ] = stiffness matrix
{q} = displacement vector
{Q} = load vector

36
3.7 Material Properties

Table 3.1 and Table 3.2 shows the material properties used for this analysis.
Soil#1 is used in the modeling of benchmark problem with no base soil (see Figure 3.7),
and Soil#2 is used in the benchmark with base soil (see Figure 3.7). These two models
are shown in Figure 3.7 and Figure 3.8.

Table 3.1: Soil properties

Material Unit weight Friction angle Cohesion Young’s Modulus Poisson’s Ratio
(kN/m3) (degree) (kN/m2)
Soil#1 20 20 10 105 0.3
5
Soil#2 20.2 19.6 3 10 0.3

Table 3.2: Reinforcement properties

Reinforcement Unit weight, γ Young’s Modulus, E Poisson’s Ratio, υ
(kg/m3) (GPa)
Nail 7,800 200 0.3
Nail head 7,800 200 0.3

3.8 Loading, and Boundary Conditions

3.8.1 Loading

In all cases, it was assumed that there is no external load other than the
gravitational forces (i.e. body force). A negative (downward) 1 kN gravity force is
applied to the whole embankment to replicate the gravity load.

37
3.8.2 Boundary Conditions
Two different geometries were used in this study: A benchmark embankment
with no base (or foundation), and a benchmark embankment with base. On the
benchmark (with no base), the bottom of the embankment is fixed and the left side to the
vertical, rollers was used as shown in Figure 3.7.

Rollers

x
Fixed Bottom

Figure 3.7: Boundary conditions used in benchmark model with no base soil

On the second model (benchmark with base soil), the bottom of the embankment
is fixed and the two sides to the vertical, rollers was used as seen in Figure 3.8.

gravity Rollers
load

Rollers

Fixed Bottom
x

Figure 3.8: Boundary conditions on benchmark model with base soil

38
3.9 Case Studies: Numerical Modeling of Unreinforced Slopes

There are two cases in this section: (a) a homogenous benchmark with no base,
and (b) a homogenous benchmark with base. The purpose of this section was to study the
effects of mesh refinement by using both linear triangular and linear quadrilateral
elements in these two cases. The mesh refinement is determined by the selection of
approximate global size.

3.9.1 Case 1: A Homogenous Slope with No Base

In the present study, a slope stability benchmark example has been considered.
This benchmark problem was solved by using different theories as explained in the
literature (Griffiths et al, 1999; Ann et al, 2004; Shiu et al, 2004) and earlier chapters.
The benchmark problem considers a uniform soil with homogeneous slope. The factors
of safety obtained from different theories are compared with the numerical results to
assess how close or far the slopes are from failure. Figure 3.9 shows the geometry of the
embankment used in the benchmark problem. In the present study, the slope of the
embankment is inclined at 26.565 degrees to the horizontal and is raised to 10 m above
the ground.

10 m Soil#1
1
2

12 m 20 m

Figure 3.9: Geometry of the benchmark model with no base model

39
 In this example, four different mesh configurations were used to study the effects
of mesh refinement on factor of safety (see Figures 3.10, 3.11, 3.12 and 3.13).

Analysis 1
6-node quadratic plane strain triangle elements were used for meshing. There
were 790 active elements and 1677 active nodes on the mesh body (see Figure 3.10).

Figure 3.10: 6-node quadratic plane strain triangle elements with approximate
global size of 0.8.

Analysis 2
8-node biquadratic plane strain quadrilateral elements were used for meshing.
There were 353 active elements and 1152 active nodes on the mesh body (see Figure
3.11).

Figure 3.11: 8-node biquadratic plane strain quadrilateral elements with

approximate global size of 0.8

40
Analysis 3
6-node quadratic plane strain triangle elements were used for meshing. There
were 2781 active elements and 5754 active nodes on the mesh body (see Figure 3.12).

Figure 3.12: 6-node quadratic plane strain triangle elements with approximate
global size of 0.4

Analysis 4
8-node biquadratic plane strain quadrilateral elements were used for meshing.
There were 1496 active elements and 7495 active nodes on the mesh body (see Figure
3.13).

Figure 3.13: 8-node biquadratic plane strain quadrilateral elements with

approximate global size of 0.4

In the analyses 1 and 2, approximate global size of 0.8 was used for meshing.
And, approximate global size of 0.4 was used in analysis 3 and 4. The value of 0.4
produces finer mesh then the size value of 0.8.

41
3.9.2 Case 2: A Homogenous Slope with Base

The second model has the same dimensions with the previous example except a
5m deep base part is added to the bottom as shown in Figure 3.14. Soil#2 properties
were used in this analysis. The geometry of the embankment is shown in Figure 3.14
below. In this model, material properties were kept same in order to have an accurate
comparison among cases that have the same geometry. The factor of safety was
computed when there was no reinforcement in the embankment. It was compared with
factor of safety with reinforcement later in this chapter.

1
2 Soil#2 15 m

5m

10 m 20 m 10 m

Figure 3.14: Geometry of the benchmark model with base soil

In this case, three different meshing systems were used to study the influence of
finite element mesh on the factor of safety. Approximate global size for mesh
refinement was varied from 0.8 to 1.5 for each analysis. Finite element meshes used in
this study are shown in Figures 3.15, 3.16, and 3.17.

42
Analysis 1
6-node quadratic plane strain triangle elements (CPE6) were used for meshing.
There were 728 active elements and 541 active nodes on the mesh body (see Figure
3.15). Approximate global size of discretization was 1.5.

Figure 3.15: 6-node quadratic plane strain triangle elements (CPE6)

Analysis 2
6-node quadratic plane strain triangle elements (CPE6) were used for meshing.
There were 1174 active elements and 1551 active nodes on the mesh body (see Figure
3.16). Approximate global size of discretization was 1.3.

Figure 3.16: 6-node quadratic plane strain triangle elements (CPE6)

43
Analysis 3
6-node quadratic plane strain triangle elements (CPE6) were used for meshing.
There were 1397 active elements and 2932 active nodes on the mesh body (see Figure
3.17). Approximate global size of discretization was 0.9.

Figure 3.17: 6-node quadratic plane strain triangle elements (CPE6)

3.10 Numerical Modeling of Reinforced Slopes

In this section, numerical modeling of reinforced slopes was studied on a

homogenous benchmark model with no base soil, and a homogenous benchmark with
base soil.

3.10.1 Influence of Soil Nail Length

Two models are analyzed in this section: a) benchmark with no soil base with one
nail and b) benchmark with soil base with one nail.

a) Benchmark with No Soil Base with One Nail

In this section, the influence of nails on stability of slopes was examined. The
factor of safety of the unreinforced slope was obtained from the previous benchmark
example. The geometry of the benchmark model and the material parameters were kept
the same in this section, in order to evaluate the soil nail influence on the factor of safety.
Deformation contours were compared and presented with and without soil nails. A nail
was placed 2m above vertically from the toe. Figure 3.18 shows the geometry of the

44
benchmark model used in this study. Nail length, L, was varied as 10m., 15m., and 20m.
Effects of the length of the nail were studied in this example.

Reinforcement
10 m
L Toe
2m

28 m 4m

Figure 3.18: Geometry of the benchmark with one soil nail.

b) Benchmark with Soil Base with One Nail

This section will examine the effects of the nail length. Similar to the previous
analysis, all the variables were kept constant except for the nail length. Nail inclination
α, was chosen as 20 degrees. A nail was placed 5m above vertically from the toe as
shown in Figure 3.18. The factor of safety was obtained when there was only one nail
with varying length. Nail length, L, was varied as 4m to 15m (see Figure 3.19).

45
 α=20°
5m 15 m
L

5m

10 m 20 m 10 m

Figure 3.19: Geometry of the benchmark with base with one soil nail.

3.10.2 Influence of Soil Nail Inclination

The study on the influence of nail inclination is presented in this section. Similar
to analysis procedure followed in previous sections, all the variables were kept constant
except the inclination angle, α. A 10m long nail was placed 5m above vertically from the
toe. Inclination was changed in two directions: clockwise inclination and
counterclockwise inclination (upward direction). The counter clockwise direction is not
very common in the field practice. In order to measure the numerical effects of soil nail
inclination, this direction was studied. Nail inclination was varied from 0 degree to 20
degree in clockwise direction and from 0 degree to 45 degree in upward direction (see
Figure 3.20).

46
 α
15 m
L=10 m
10 m
5m

10 m 20 m 10 m
Figure 3.20: Geometry of the benchmark with base with inclined nail

3.10.3 Influence of Soil Nail Location

In this section the effects of location of soil nail is examined. A 10m long nail was
placed at different locations (Figure 3.21) to study the effects of the nail location. One
nail was used in each simulation. Along with small increments, the nail was placed
starting from the toe level of the slope to 9m above vertically from toe. Similar to the
previous examples, all the variables were kept constant except the location of the nail.

L
At the toe
L
L 15 m
L
5m L

10 m 20 m 10 m

Figure 3.21: Geometry of the benchmark with base with different locations

47
3.10.4 Influence of Multiple Nails

In addition to the single nail study, multiple nails were placed into the slope
surface as shown in Figure 3.22. In this section, the number of nails was increased from
one to three to examine the effects of multiple nails. The length of the nails was chosen as
10 m. The vertical spacing of the nails was 2 m and the horizontal spacing was 4 m (see
Figure 3.22). The material properties of soil, nails and nail heads were the same as
previous examples. Single nail case was compared with two-nail and three-nail case in
terms of the factor of safety values.

Results from all these cases are shown in the next chapter.

θ = 26.565°
2m α = 20°
L 10 m
2m
L
5m L

5m 4m 4m 19 m

Figure 3.22: Geometry of the benchmark with base with multiple nails

48
 CHAPTER 4

RESULTS and DISCUSSIONS

4.1 Introduction

In the present study, two-dimensional finite element analyses were performed

using ABAQUS to evaluate stability of slopes with and without reinforcement. Analyses
were performed on two different slopes (geometric models) with several different cases
as described in the previous chapter. The factors of safety for each analysis were obtained
from different theories and previous research work reported in the literature (Griffiths and
Lane, 1999; Ann et al, 2004; Shiu and Chang, 2004). The results are compared with the
numerical results to evaluate the suitability of numerical methods to analyze slope
failures. Also, the results obtained from ABAQUS are compared with results obtained
from other finite element schemes reported in the literature (Griffiths and Lane, 1999). In
finite element analyses, the factor of safety is determined by using the Shear Strength
Reduction (SSR) technique which was explained in previous chapters.

4.2 Results for Unreinforced Slopes

The results are shown for the following cases in this section: a homogenous slope
model with no base soil, and a homogenous slope with base soil.

4.2.1 Case 1: A Homogenous Slope with No Base

In this example, four different meshing systems were used to study the effects of
concentration of mesh refinement on factor of safety. The results are shown in Table 4.1
from the analyses 1, 2, 3, and 4. The computed factor of safety ranged from 1.32 to 1.401
as shown in Table 4.1. These minimum and maximum values are less than 3% away
from those obtained from limit equilibrium methods. However, the difference of
computed factor of safety using CPE6 and CPE8 elements are slightly larger than 5%.
As would be expected, the finer mesh gives more conservative results than the coarser
mesh. The factor of safety using CPE6 and CPE8 from Analyses 1 and 2, are very close
to the results reported by Griffiths (Griffiths et. al. 1999).

49
 The deformed finite element meshes corresponding to these cases are shown in
Figures 4.1(a), 4.2 (a), 4.3 (a), and 4.4(a). The equivalent plastic strain measure in finite
element analysis can be used to assess the accumulation of plastic strain in the soil region
(Abaqus, 2006). Figures 4.1 (b), 4.2 (b), 4.3 (b), and 4.4 (b) show contours of the
equivalent plastic strain for Case 1. Additionally, a 5m base was added to this slope to
investigate the influence of soil base on the factor of safety (see Appendix A). The
computed factor of safety was found as 1.395.

Table 4.1: Computed Factor of safety for Case 1

Percentage difference in
Metho Factor of FS compared to Griffiths’
Case 1 d Mesh Safety results
Analysis 1 (790 elements) FE CPE6 1.401 -0.07%
Analysis 2 (353 elements) FE CPE8 1.385 1.07%
Analysis 3 (2781 elements) FE CPE6 1.341 4.21%
Analysis 4 (1496 elements) FE CPE8 1.329 5.07%
Griffiths' (1999) FE Q8 1.400 -
Janbu (Slope/W) LEM - 1.390 0.71%
Bishop (Slope/W) LEM - 1.382 1.29%
M-P (Slope/W) LEM - 1.392 0.57%
*FS= Factor of Safety
*M-P= Morgenstern-Price Method

Figure 4.1 (a): Deformed mesh with CPE6 elements (Analysis 1)

50
Figure 4.1 (b): The equivalent plastic strain contour for Analysis 1

Figure 4.2 (a): Deformed mesh with CPE8 elements (Analysis 2)

Figure 4.2 (b): The equivalent plastic strain contour for Analysis 2

51
Figure 4.3 (a): Deformed mesh with CPE6 elements (Analysis 3)

Figure 4.3 (b): The equivalent plastic strain contour for Analysis 3

Figure 4.4 (a): Deformed mesh with CPE8 elements (Analysis 4)

52
 Figure 4.4 (b): The equivalent plastic strain contour for Analysis 4

The computed factors of safety were close to limit equilibrium values (<5%),
finer mesh with CPE8 elements performed well on replicating the failure surface on
deformed mesh (Figure 4.4 a). Moreover, CPE6 elements performed better than CPE8
elements on the shape of failure plane. However, more research is needed to confirm this
observation. The effects of the proper element selection for slope stability analyses can be
a research topic itself.

4.2.2 Case 2: A Homogenous Slope with Base

Finite element analysis of a slope with a base was performed by using three
different mesh refinements were used to study the effects of meshing on the factor of
safety. In this example, only CPE6 elements were used in all three cases. Three meshes
used in example had following properties: analysis 1 with 728 active elements (Figure
4.5), analysis 2 with 1164 active elements (Figure 4.6), and analysis 3 with 1397 active
elements (Figure 4.7). The results on the computed factor of safety are shown in Table
4.2. Consistently, the finer mesh gave more conservative results than the coarser mesh.

The factor of safety ranged from 0.98 to 1.03 in this case. These minimum and
maximum values are less than 5% away from the values based on limit equilibrium
methods (Bishop’s Modified Method (Bishop, 1955), Janbu’s Modified Method (Janbu,
1968), Morgenstern and Price’s Method (Morgenstern and Price, 1965), and Spencer’s
Method (Spencer, 1967)). However, the difference in computed factor of safety using 728
elements (Analysis 1) and 1397 elements (Analysis 3) elements is slightly larger than 5%.

53
Table 4.2: Computed Factor of safety results for Case 2
Percentage difference
Factor of in FS compared to min.
Case 2 Method Mesh Safety LEM results (0.989)
Analysis 1 (728 Elements) FEM CPE6 1.03 4.15%
Analysis 2 (1,164 Elements) FEM CPE6 0.995 0.61%
Analysis 3 (1,397 Elements) FEM CPE6 0.98 -0.91%
Spencer (Slope/W) LEM - 0.989 -
Bishop (Slope/W) LEM - 0.99 -
Morgenstern-Price (Slope/W) LEM - 0.999 -

The deformed finite element meshes corresponding to these three cases are shown
in Figures 4.5 (a), 4.6 (a), and 4.7 (a). The deformed mesh shows the shape of the sliding
block as shown in figures. The equivalent plastic strain measure in finite element
analysis can be used to assess the accumulation of plastic strain in the soil region
(Abaqus, 2006). Figures 4.5 (b), 4.6 (b), and 4.7 (b) show contours of the equivalent
plastic strain for the one-nail case.

Figure 4.5 (a): Deformed mesh with 728 elements (Analysis 1)

54
Figure 4.5 (b): The equivalent plastic strain contour for Analysis 1

Figure 4.6 (a): Deformed mesh with 1164 elements (Analysis 2)

Figure 4.6 (b): The equivalent plastic strain contour for Analysis 2

55
 Figure 4.7 (a): Deformed mesh with 1397 elements (Analysis 3)

Figure 4.7 (b): The equivalent plastic strain contour for Analysis 3

4.2.3 The Radius of failure circle compare with Limit Equilibrium Method

For this analysis, same slope from the Analysis 1 in Section 4.2.1 example has
been considered. The benchmark problem considers a uniform soil with homogeneous
slope. The computer software package GeoStudio 2004-SLOPE/W developed by GEO-
SLOPE was used in determining the factor of safety values are obtained from the limit
equilibrium methods that were mentioned in Chapter 2.

The finite element result of failure circle of the slope for Analysis 1 in this section
was compared with the limit equilibrium method results. The radius point was obtained
by using Geo-Slope (Figure 4.8). For this example, the radius was 20.33 m, and the
coordinates were found to be x=27.34 m and y=20.16 m. The same coordinates were

56
sketched on the finite element analysis of slope displacement contours (Figure 4.9 (a) and
the equivalent plastic strain contours (Figure 4.9 (b)).

y (m)

Radius point
(27.34, 20.16)

R=20.33 m

Figure 4. 8: Limit equilibrium solution for Analysis 1 generated with Geo Studio,
Slope/W (Geo-Studio, 2004)

57
 Radius point
(27.34, 20.16)
y (m)

R= 20.33 m

10

0 32 x (m)

Figure 4.9 (a): Displacement contours for Analysis 1

58
 Radius point
(27.34, 20.16)
y (m)

R=20.33 m

10

0 32 x (m)

Figure 4.9 (b): The equivalent plastic strain contours for Analysis 1

The finite element analysis displacement and the equivalent plastic strain contours
of the homogeneous slope replicated the radius of failure circle of the limit
equilibrium method results. The finite element method was able to predict the failure
plane (slip surface) on both displacement contours (Figure 4.9 (a)) and the equivalent
plastic strain contours (Figure 4.9 (b)). In addition, the finite element analysis in this
example did not require an arbitrary partitioning of the critical surface selection prior
to calculations.

4.3 Results for Reinforced Slopes

The results are shown for the following cases in this section: a homogenous
benchmark model with no base soil, and a homogenous benchmark with base soil.

59 x (m)
4.3.1 Influence of Soil Nail Length

The effects of the length of a nail were studied in this section: a) benchmark with
no soil base with one nail and b) benchmark with soil base with one nail.

a) Benchmark with No Soil Base with One Nail

In this case, the effect of the length of a nail was studied on three models as
described in section 3.10.1 in the previous chapter. The length of the nail affected the
slope deformation significantly. Increasing the nail length beyond a certain length did
not change the displacements significantly. For the case with a 10 m long nail, the nail
was not long enough to pass the failure plane, and therefore it did not improve the
deformation as much (Figure 4.10 (a)). When the slope did not have any reinforcement
the obtained factor of safety was 1.40. After the addition of the 10 m long nail, the factor
of safety was increased by about 4.3% and it was determined as 1.46. In the second case,
a 15 m long nail was used. This nail resulted in a significantly higher factor of safety,
1.58. This is about 12.86% increase comparing to the case of an unreinforced slope. The
results are shown in Table 4.3. Deformed meshes corresponding to different nail lengths
are shown in Figures 4.10, 4.11, and 4.12. As can be seen from these figures, the failure
surface is different for reinforced cases in comparison to the unreinforced case.

Figure 4.10 (a): Deformed mesh for the case with a 10 m long nail

60
Figure 4.10 (b): Displacement contours for the case with a 10 m long nail

Figure 4.11 (a): Deformed mesh for the case with a 15 m long nail

Figure 4.11 (b): Displacement contours for the case with a 15 m long nail

61
 Figure 4.12 (a): Deformed mesh for the case with a 20 m long nail

Figure 4.12 (b): Displacement contours for the case with a 20 m long nail

Table 4.3: Computed Factor of safety for unreinforced slope and one-nail case

Unreinforced Case 1 with one nail

Slope Nail Length (m)
10m 15m 20m
1.40 1.46 1.52 1.59

62
b) Benchmark with Soil Base with One Nail

In this case, the influence of the nail length was examined as described in Section
3.10.1. All the variables were kept constant except for the nail length. Nail inclination α,
was chosen as 20° with respect to the geometry defined earlier (see Figure 3.20). A nail
was placed 5m vertically above the toe. The factor of safety was obtained when there
was only one nail, but with varying length. Nail length, L, was varied as 4m to 15m. The
greatest factor of safety of 1.33 was obtained when the soil nail length had a length of
12m. Variation of factor of safety with nail length is shown in Figure 4.13. As can be
seen from this Figure, the critical length of the nail appears to be about 10m. Beyond this
length, no improvements were observed.

1.35

1.3

1.25
Factor of Safety

1.2

1.15

1.1

1.05

1
4 5 6 7 8 9 10 11 12 13 14 15
Length (m)

Figure 4.13: Variation of factor of safety against length of the soil nail

The resultant displacement was determined at point “A” for different nail length
as shown in Figure 4.12 to investigate the relationship between nodal displacement and
the factor of safety value. Variation of displacement at point ‘A’ against length of the soil
nail is shown in Figure 4.14. The results show that the decrease in displacement values
correlate with increase in the nail length values. This figure shows that there was

63
significant difference in displacement when the length of the nail falls between 5m and
9m. Figure 4.15 shows that the factor of safety starts to become stable after the nail
length exceeded 9m.

Point A

Figure 4.14: Schematic location of point “A”

0.2

0
0 2 4 6 8 10 12 14 16
-0.2

-0.4
Displacement (m)

-0.6

-0.8

-1

-1.2

-1.4

-1.6

-1.8
Nail Length (m)

Figure 4.15: Variation of displacement at point ‘A’ with the length of soil nail

64
 Plastic strain contours corresponding to different values of nail length are shown
in Figures 4.16 to 4.19. The plastic strain values in different regions in the soil and
additional information on the proximity of the existing shear stress to the maximum shear
strength that can be generated in those regions can be helpful in improving designs for
reinforced soil slopes (Abaqus, 2006). Slope failures can occur when adjacent regions of
the soil mass yield simultaneously in shear. The equivalent plastic strain measure in finite
element analysis can be used to assess the accumulation of plastic strain in the soil region
(Abaqus, 2006). Figures 4.16, 4.17, 4.18, and 4.19 show contours of the equivalent
plastic strain for the one-nail case.

Figure 4.16: The equivalent plastic strain contour for 4m long nail

65
Figure 4.17: The equivalent plastic strain contour for 6m long nail

Figure 4.18: The equivalent plastic strain contour for 8m long nail

66
 Figure 4.19: The equivalent plastic strain contour for 15m long nail

4.3.2 Influence of Nail Inclination

Nail inclination was varied from 0° to 26° degree in clockwise (positive) direction
and from 0° degree to 45° degree in counterclockwise (negative) direction. A 10m long
nail was placed 5m vertically above from the toe as shown in Figure 4.19. Similar to the
previous examples, all the variables were kept constant except for the nail inclination
angle. The highest calculated factor of safety value of 1.33 was obtained when the
inclination angle was at 20° in clockwise direction. Nail length, L, was varied from 4m
to 15m. The relationship between the calculated factor of safety and the inclination angle
α is presented in Figure 4.20. The factor of safety is close to 1.30 with small variations
for the range of α between 0° and 20° in the clockwise direction. The factor of safety
decreased considerably as the inclination angle α increased beyond 20°.

67
 1.40

1.20

Factor of Safety 1.00

0.80

0.60

0.40

0.20

0.00
-50 -40 -30 -20 -10 0 10 20 30
Nail Inclination

Figure 4.20: Variation of factor of safety with inclination angle of a soil nail

Figures 4.21 to 4.23 show the failure mechanism of the slope for different values
of nail inclination angle.

Figure 4.21: The equivalent plastic strain contours for nail inclination of zero
degrees

68
Figure 4.22: The equivalent plastic strain contours for nail inclination of 20
degrees (counterclockwise)

Figure 4.23: The equivalent plastic strain contours for nail inclination of 45
degrees

69
4.3.3 Influence of Nail Location

In this section, the influence of location of soil nail is presented as described in

Section 3.10.3. A 10m long nail was placed in different locations to study the effects of
the nail location. One nail was used in each simulation. The nail location was changed
from the toe level of the slope to 9m vertically above from toe (Figure 4.26). The results
are shown in the Figure 4.24. The maximum factor of safety value of 1.28 was obtained
when the nail was at 5.5 m above the toe level. The factor of safety values started to
decrease beyond this position.

The failure envelope is clearly seen in Figure 4.25 which shows plastic strain
contours; this figure corresponds to the case in which the nail was located at the toe of the
slope. Figure 4.24 corresponds to the case in which the nail was located 8m above the
toe. Both these figures show that plastic strain contours can be used to identify the
failure surface.

1.4

1.2

1
Factor of Safety

0.8

0.6

0.4

0.2

0
0 1 2 3 4 5 6 7 8 9 10
Vertical Location (m)

Figure 4.24: Variation of factor of safety with nail location

70
Figure 4.25: The equivalent plastic strain contours for the case in which nail was
located at the toe.

Figure 4.26: The equivalent plastic strain contours for one nail at 9m above from
the toe of slope

71
4.3.4 Influence of Number of Nails

In this section, the number of nails was increased from one to three to examine the
influence of number of nails on the factor of safety. The unreinforced case and the case
with one nail were studied in earlier cases (Section 3.10.4). After the addition of the
second nail, the factor of safety increased by about 3.5%; it was determined as 1.154
compared to the case with one nail. When the third nail was introduced in the slope, the
factor of safety of the slope increased significantly, up by about 22%. This is a very large
increase compared to the unreinforced, and single-nail slope cases. The variation of
factor of safety values are shown in Table 4.4.

Table 4.4: Factor of safety results for influence of number of nails on the factor of safety

Unreinforced Case 2 with soil nails

Slope
One-Nail Two-Nails Three-Nails

0.995 1.10 1.154 1.406

Figures 4.27 and 4.28 show that the failure plane in the three-nail model was raised above
the nails when compared to single-nail model.

72
Figure 4.27: The equivalent plastic strain contour for one nail at the toe of slope

Figure 4.28: Displacement contour of three-nail case

73
 CHAPTER 5

SUMMARY AND CONCLUSIONS

5.1 Summary

The factor of safety of slope stability was determined by using the finite element
method in conjunction with the Shear Strength Reduction (SSR) technique which was
explained in previous chapters. The primary focus of this research was to study: (a) the
influence of meshing and soil nailing on the factor of safety values, and (b) the failure
mechanism of slopes by using the finite element analysis. Seven cases were studied in
this research as summarized below. Summary of results and findings are presented in
Table 5.1 and 5.2.

Table 5.1: Factor of safety results for homogeneous slope with no base (Case 1)

Factor of Safety
Case 1 Limit Equilibrium Method Finite Element Method
Unreinforced min. 1.382 (with Bishop) min. 1.341 (with 1,496 CPE8 Elements)
Slope max. 1.392 (with M-P*) max. 1.401 (with 790 CPE6 Elements)
min. 1.46 (when L=10m)
Reinforced Slope max. 1.58 (when L=20m)
with one-nail The failure plane of the slope was raised
(Figure 3.18) above the nail (Figure 4.12)

*M-P: Morgenstern-Price Method (1965)

74
Table 5.2: Factor of safety results for homogeneous slope with base (Case 2)

Factor of Safety
Case 2 Limit Equilibrium Method Finite Element Method
Unreinforced min. 0.989 (with Spencer) min. 0.980 (with 1,397 Elements (CPE6))
Slope max. 0.999 (with M-P*) max. 1.03 (with 728 Elements (CPE6))
Reinforced
Slope
Influence of Nail min. 1.07 (when L=4m)
Length max. 1.33 (when L=10m)
Influence of Nail min. 1.01 (when α=20° counterclockwise)
Inclination max. 1.33 (when α=20° clockwise)
Influence of Nail min. 1.02 (when vertical distance above toe level= 0 m)
Location max. 1.28 (when vertical distance above toe level= 5.5 m)
min. 1.46 (when L=10m)
max. 1.58 (when L=20m)
Influence of The failure plane of the slope was raised above the last
Number of Nails nail (Figures 4.27 and 4.28)

*M-P: Morgenstern-Price Method (1965)

5.2 Numerical Modeling of Unreinforced Slopes

In this section, numerical modeling of unreinforced slopes was studied on a

homogenous slope model with no base soil, and a homogenous slope with base soil.

5.2.1 Case 1: A Homogenous Slope with No Base

In this case, 6-noded quadratic plane strain triangle elements (CPE6) elements and
8-noded biquadratic plane strain quadrilateral elements (CPE8) elements were used in
section. Results show that the minimum and maximum values of factor of safety are less
than 3% of the limit equilibrium values. However, the difference of computed factor of
safety using CPE6 elements and CPE8 elements are slightly larger than 5%. The factor
of safety using CPE6 and CPE8 from Analyses 1 and 2 are very close to the results
presented in the literature for Q8 elements (Griffiths et. al.). Although, the computed
factors of safety were close to limit equilibrium values, finer mesh with CPE8 elements
did not perform well in replicating the failure surface. Moreover, CPE6 elements
performed better than CPE8 elements on the shape of failure plane.

75
5.2.2 Case 2: A Homogenous Slope with Base

In this case, only CPE6 elements were used in all three models. The computed
factor of safety using ranged from 0.98 to 1.03 in this case. These minimum and
maximum values are less than 5% from the limit equilibrium values. The limit
equilibrium methods lead to a factor of safety value in the range of 0.989 to 0.999 (see
Table 4.1). However, the difference of computed factor of safety using 728 elements
(Analysis 1) and 1397 elements (Analysis 3) elements are slightly larger than 5%. This
shows that the finer refined mesh system could lead more conservative predictions.

5.3 Numerical Modeling of Reinforced Slopes

In this section, numerical modeling of reinforced slopes was studied on a

homogenous slope model with no base soil, and a homogenous slope with base soil.

5.3.1 Influence of Soil Nail Length

Two cases are analyzed in this section: a) benchmark with no soil base with one
nail and b) benchmark with soil base with one nail.

a) Benchmark with No Soil Base with One Nail

When the slope did not have any reinforcement, the computed factor of safety was
1.40. After the addition of a 10 m long nail, the factor of safety was increased by about
4.3% and it was determined as 1.46. In the second case, a 15 m long nail was used. This
trial resulted in a significantly higher factor of safety value of 1.58. This is about 12.86%
increase compared to the unreinforced slope case. Clearly, the finite element method is
capable of computing the factor of safety of reinforced slopes. Traditional methods have
not been used with reinforced slopes.

76
b) Benchmark with Soil Base with One Nail

The greatest factor of safety 1.33 was obtained when the soil nail length was 12m.
There was a significant difference in displacement when the length of the nail was
between 5m and 9m. The results show that the slope starts to become stable after a soil
nail with a length of 9m to 12 m was used (Figure 4.13).

5.3.2 Influence of Nail Inclination

The highest calculated numerical value factor of safety of 1.33 was obtained when
the nail inclination angle (α) was at 20° in clockwise direction. The factor of safety was
close to 1.30 with small variations for the range of α between 0° and 20° in the clockwise
direction. The factor of safety decreased considerably as the inclination angle α
increased beyond 20°.

5.3.3 Influence of Nail Location

The maximum factor of safety value of 1.28 was obtained when the nail was
installed at 5.5 m above the toe level. The factor of safety values started to decrease
beyond this height.

5.3.4 Influence of Number of Nails

The factor of safety increased by about 3.5% after addition of the second nail. It
was determined as 1.154 compared to one-nail case. When the third nail was introduced
in the slope, the factor of safety of the slope increased significantly 1.406 (up to about
22%). The failure plane of the slope was raised above the last nail.

77
5.4 Conclusions

Following conclusions were drawn based on the results presented in this research
work:

• The primary focus of this research was to obtain factor of safety of earth slopes by
finite element analysis. The performance of the finite element method for
determining factor of safety and the failure surface in conjunction with the
strength reduction method was excellent. Obtaining the factor of safety of a soil
slope by finite element analysis is an effective method for slope failure
investigations.

• Soil nailing is an effective slope stabilization method. The finite element method
can predict the factor of safety and the failure plane on reinforced slopes.
Conventional Limit Equilibrium methods have not been used for reinforced
slopes.

• Finer mesh refinement could lead more conservative predictions.

• The location of the critical slip plane of a slope is accurately determined by using
the finite element analysis. Unlike the traditional methods, finite element method
does not require an arbitrary partitioning of the critical surface selection prior to
calculations.

• The location of soil nails is an important factor on slope stability. Placing the first
nail at the toe of a slope is less effective then placing the nail in the middle.

• The inclination of a soil nail has a significant influence on the factor of safety of
an earth slope and the extent of the critical slip surface. Finite element analysis is
an accurate method to determine the optimal inclination amount of a soil nail for a
given slope geometry and soil properties.

78
5.5 Recommendations

• The validity of the finite element analysis results for both unreinforced and
reinforced have been compared with other analysis methods; however, field data
and case studies are still needed to further calibrate the method.

• Soil nails have a significant importance in soil stabilizing. However, soil nail and
nail head design were out of scoop of this project. Therefore, soil nail and nail
head design with finite element analysis can be the subject of a future study.

• The finite element prediction using the Mohr-Coulomb model was acceptable.
However, it is recommended to use advanced soil models such as a strain-
hardening soil model with a laboratory testing program to achieve more reliable
predictions.

• The additional research is needed to investigate slope stability with broader range
of soils and element types.

79
 REFERENCES

ABAQUS version 6.6 (2006) ABAQUS user’s manual. Hibbitt, Karlson & Sorensen, Inc.,
Pawtucket, RI.

Ann,T., Hai, O., Lum, C., and Tan, D. (2004), Finite Element Analysis of A Soil Nailed
Slope-Some Recent Experience. pp. 183-189

ASTM A615/A615M-96 (1996) Standard Specification for Deformed and Plain Billet-
Steel Bars for Concrete Reinforcement. American Standard for Testing Materials.

Baker, R. (2004) Nonlinear Mohr Envelopes Based on Triaxial Data. Journal of

Geotechnical & Geoenvironmental Engineering, Vol. 130 Issue 5, pp. 498-506

Banerjee, S., Finney, A., Wentworth, T. and Bahiradhan, M. (1998) Evaluation of

Design Methodologies for Soil-Nailed Walls. Volume 3: An Evaluation of Soil-
Nailing Analysis Packages,” FHWA Publication No. WA-RD 371.1, pp. 154

Bishop, A.W. (1955) The Use of the Slip Circle in the Stability Analysis of Slopes.
Géotechnique, London, England, Vol. 5, No. 1, pp. 7-17

Bromhead, E. N. (1992) Stability of Slopes. Second Edition, Blackie Academic &

Professional

Broms, B.B. and Wong, K.S. (1991) Landslides in Fang, H., ed., Foundation
Engineering Handbook: New York, Van Nostrand Reinhold, pp. 410-446.

Byrne, R. J., Cotton, D., Porterfield, J., Wolschlag, C., and Ueblacker, G. (1998) Manual
for Design and Construction Monitoring of Soil Nail Walls. Report FHWA-SA-
96-69R, Federal Highway Administration, Washington, D.C.

Chance Company, Retention Wall System Design Manual [Online], (1999) Available:
http://www.abchance.com/ch_tech/soilscrew_designman/chapter1.htm#anchor11
14753

Chen, J. Lin, C. and Lee, W. (2003) Dilative Behavior of Granular Material

International Journal of Offshore and Polar Engineering Vol. 13, No. 4

Chen, W. R. (1995) The Civil Engineering Handbook. Editor-in-Chief, CRC Press, Inc.,
Boca Raton, Florida, pp. 774-798.

Das, B. M. (2002) Principles of Geotechnical Engineering. Fifth Edition. Pacific Grove,

CA: Brooks/Cole.

80
Dawson, E. M., Roth, W.H. and Drescher, A. (1999) Slope Stability Analysis by Strength
Reduction. Geotechnique, Vol. 49, No. 6, pp. 835-840

Duncan, J.M. and Wright, S.G. (2005) Soil Strength and Slope Stabilty. John Wiley &
Sons, Inc., pp. 199

Fenton, G.A. (1990) Simulation and Analysis of Random Fields. thesis presented to
Princeton University, at Princeton, New Jersey, in partial fulfillment of the requirements
for the degree of Doctor of Philosophy, 178 pg., Jan. 1990.

FHWA, (1993). French National Research Project CLOUTERRE, 1991-

Recommendations CLOUTERRE 1991, (English Translation) Soil Nailing
Recommendations, Publication FHWA-SA-93-026, Federal Highway
Administration, Washington, D.C.

GeoStudio 2004, version 6.2, Student Edition, Geo-Slope International Ltd.

Goldstein, N. (2001) [Online] “Soil Nailing” Available:

http://www.forester.net/ec_0109_soil.html

Griffiths, D.V., and Lane P.A. (1999) Slope Stability Analysis by Finite Elements.
Geotechnique, Vol. 49, No. 3, pp. 387-403

Hammah, R.E., Curran, J.H., Yacoub T., Corkum, B. (2004) Stability Analysis of Rock
Slopes using the Finite Element Method. EUROCK 2004 & 53rd Geomechanics
Colloquium. Schubert (ed.)

Holtz, R.D, and Kovacs, W.D. (1981) An Introduction to Geotechnical Engineering.

Prentice-Hall, Inc. pp. 448

Joshi, B. (2003) Behavior of Calculated Nail Head Strength in Soil-Nailed Structures

Journal of Geotechnical and Geoenvironemtal Engineering, Vol. 129, Issue 9, pp.
819-828

Karakus, M. and Fowell, R.J. (2004) An insight into the New Austrian Tunnelling Method
(NATM), ROCKMEC′2004-VIIth Regional Rock Mechanics Symposium,
Sivas, Turkey

KUTC, Newsletter, Spring (2006) Kansas Local Technical Assistance Program (LTAP)
at the University of Kansas Transportation Center

Morgenstern, N.R. and Price, V.E. (1965) The analysis of the stability of general slip
Surfaces, Géotechnique, Vol. 15, No. 1, pp. 79-93

Rocscience (2004) A New Era in Slope Stability Analysis: Shear Strength Reduction
Finite Element Technique, Article

81
Shiu, Y. K., Chang G.W.K., and Cheung, W.M. (2006) Review of Limit Equilibrium
Methods for Soil Nail Design, Special Project Report SPR3, Geotechnical
Engineering Office, Hong Kong

Shiu, Y. K., and Chang G.W.K. (2004) Numerical Study on Soil Nail Heads and
Inclination, Geotechnical Engineering Office, Civil Engineering and
Development Department, HKSAR, China

Shong, L.S. (2005) Soil Nailing for Slope Strengthening, Geotechnical Engineering, pp.
1-9

Spencer, E. (1967) A Method of Analysis of the Stability of Embankments Assuming

Parallel Interslice Forces, Géotechnique, Vol. 17, No. 1, pp. 11-26

Terzaghi, K., Peck, R.B., and Mesri, G. (1996) Soil Mechanics in Engineering Practice.
Third Edition, John Wiley & Sons, Inc. Article 18, pp. 135

U.S. Army Corps of Engineers, (2003) Engineer Manual. [Online] Available:

http://www.usace.army.mil/publications/eng-manuals/em1110-2-1902/entire.pdf

Williams Form Engineering Corporation [Online], (2006) Available:

http://www.williamsform.com/Threaded_Bars/Grade_75_All-
Thread_Rebar/grade_75_all-thread_rebar.html

Yu, H. S. (2000) Cavity Expansion Methods in Geomechanics, Kluwer Academic

Publishers, pp. 404

82
 APPENDIX A

Influence of Soil Base on Factor of Safety

In this section, a base was added to Case 1 model. Figure A 2 shows the
additional 5m base and a 10m stretch to the toe section to study the failure surface.
Soil#1 properties were used in both analyses. In this model, material properties were kept
same in order to have an accurate comparison between these models. The factor of safety
was computed when there was no reinforcement in the embankment.

10 m Soil#1
1
2

12 m 20 m

Figure A 1: Geometry of the benchmark model with no base model (Case 1)

Soil#1
1
15 m 2

5m

12 m 20 m 10 m

Figure A 2: Geometry of the benchmark model with a 5m deep base added

83
 The deformed finite element meshes corresponding to these cases are shown in
Figures A 1 and A 2. When the slope did not have a base (Figure A 1), the obtained
factor of safety was 1.40. After the addition of the 5 m base (Figure A 2), the factor of
safety was determined as 1.395.

Figure A 3: Deformed mesh for Case 1 (from Analysis 1)

Figure A 4: Deformed mesh for Case 1 with base

The equivalent plastic strain measure in finite element analysis can be used to
assess the accumulation of plastic strain in the soil region (Abaqus, 2006). Figure A 5
shows displacement and Figure A 6 shows contours of the equivalent plastic strain for
Case 1 with a 5m base. Figure 4.1 (b) in Chapter 4 shows the equivalent plastic strain
contour for Case 1.

84
Figure A 5: Displacement contours

Figure A 6: The equivalent plastic strain contours

85
Summary

Table A 1 shows the computed factor of safety for Case 1 and Case 1 with a 5m deep
base added. The computed factor of safety values for Case 1 and Case 1 with base are
shown in Table A.1. The error percentage is less than 0.4%.

Table A1: Computed Factor of safety for Case 1 and Case 1 with a 5m deep base added

Case 1 Case 1 with Base Error percentage

Factor of Safety 1.4 1.395 0.36%

86